GAIA
Meta's benchmark of real-world questions testing reasoning, multimodality, web browsing, and tool-use for general assistants.
GAIA scores 2.6/10 (Nascent) in Benchmarks & Resources.
Kind Resource License No license (all rights reserved) Repo gaia-benchmark/leaderboard
Quality
Nascent 2.6
Adoption 0
Activity 35
Maturity 0
Community 0
Capability 50
Show the math
Overall 2.6/10
value = 2.6/10
score = value
Adoption 1/10
value = 0/100
score = 1 + 9 * value / 100
Activity 4.2/10
value = 35/100
score = 1 + 9 * value / 100
Maturity 1/10
value = 0/100
score = 1 + 9 * value / 100
Community 1/10
value = 0/100
score = 1 + 9 * value / 100
Capability 5.5/10
value = 50/100
score = 1 + 9 * value / 100
Key metrics
GitHub Repository
Gotchas
No gotchas documented yet. Contribute one if you know a constraint we missed.
Labels
- No license
No OSS license
No recognized open-source license is declared; using it commercially is legally ambiguous.
benchmark reasoning tool-use evaluation